blob: 7b99636564c8cbc608ff6fb1a24b883c254bf8e3 [file] [log] [blame]
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -08001Documentation for userland software suspend interface
2 (C) 2006 Rafael J. Wysocki <rjw@sisk.pl>
3
4First, the warnings at the beginning of swsusp.txt still apply.
5
6Second, you should read the FAQ in swsusp.txt _now_ if you have not
7done it already.
8
9Now, to use the userland interface for software suspend you need special
10utilities that will read/write the system memory snapshot from/to the
11kernel. Such utilities are available, for example, from
Rafael J. Wysockibf73bae2006-12-06 20:34:16 -080012<http://suspend.sourceforge.net>. You may want to have a look at them if you
13are going to develop your own suspend/resume utilities.
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -080014
15The interface consists of a character device providing the open(),
16release(), read(), and write() operations as well as several ioctl()
Rafael J. Wysocki3010f8c2007-10-26 01:05:05 +020017commands defined in include/linux/suspend_ioctls.h . The major and minor
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -080018numbers of the device are, respectively, 10 and 231, and they can
19be read from /sys/class/misc/snapshot/dev.
20
21The device can be open either for reading or for writing. If open for
22reading, it is considered to be in the suspend mode. Otherwise it is
Rafael J. Wysockibf73bae2006-12-06 20:34:16 -080023assumed to be in the resume mode. The device cannot be open for simultaneous
24reading and writing. It is also impossible to have the device open more than
25once at a time.
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -080026
27The ioctl() commands recognized by the device are:
28
29SNAPSHOT_FREEZE - freeze user space processes (the current process is
Rafael J. Wysockicc5d2072007-10-26 01:03:33 +020030 not frozen); this is required for SNAPSHOT_CREATE_IMAGE
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -080031 and SNAPSHOT_ATOMIC_RESTORE to succeed
32
33SNAPSHOT_UNFREEZE - thaw user space processes frozen by SNAPSHOT_FREEZE
34
Rafael J. Wysockicc5d2072007-10-26 01:03:33 +020035SNAPSHOT_CREATE_IMAGE - create a snapshot of the system memory; the
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -080036 last argument of ioctl() should be a pointer to an int variable,
37 the value of which will indicate whether the call returned after
38 creating the snapshot (1) or after restoring the system memory state
39 from it (0) (after resume the system finds itself finishing the
Rafael J. Wysockicc5d2072007-10-26 01:03:33 +020040 SNAPSHOT_CREATE_IMAGE ioctl() again); after the snapshot
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -080041 has been created the read() operation can be used to transfer
42 it out of the kernel
43
44SNAPSHOT_ATOMIC_RESTORE - restore the system memory state from the
45 uploaded snapshot image; before calling it you should transfer
46 the system memory snapshot back to the kernel using the write()
47 operation; this call will not succeed if the snapshot
48 image is not available to the kernel
49
50SNAPSHOT_FREE - free memory allocated for the snapshot image
51
Rafael J. Wysockicc5d2072007-10-26 01:03:33 +020052SNAPSHOT_PREF_IMAGE_SIZE - set the preferred maximum size of the image
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -080053 (the kernel will do its best to ensure the image size will not exceed
54 this number, but if it turns out to be impossible, the kernel will
55 create the smallest image possible)
56
Rafael J. Wysockiaf508b32007-10-26 00:59:31 +020057SNAPSHOT_GET_IMAGE_SIZE - return the actual size of the hibernation image
58
Rafael J. Wysockicc5d2072007-10-26 01:03:33 +020059SNAPSHOT_AVAIL_SWAP_SIZE - return the amount of available swap in bytes (the
60 last argument should be a pointer to an unsigned int variable that will
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -080061 contain the result if the call is successful).
62
Rafael J. Wysockicc5d2072007-10-26 01:03:33 +020063SNAPSHOT_ALLOC_SWAP_PAGE - allocate a swap page from the resume partition
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -080064 (the last argument should be a pointer to a loff_t variable that
65 will contain the swap page offset if the call is successful)
66
Rafael J. Wysockicc5d2072007-10-26 01:03:33 +020067SNAPSHOT_FREE_SWAP_PAGES - free all swap pages allocated by
68 SNAPSHOT_ALLOC_SWAP_PAGE
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -080069
Rafael J. Wysockibf73bae2006-12-06 20:34:16 -080070SNAPSHOT_SET_SWAP_AREA - set the resume partition and the offset (in <PAGE_SIZE>
71 units) from the beginning of the partition at which the swap header is
72 located (the last ioctl() argument should point to a struct
Rafael J. Wysocki3010f8c2007-10-26 01:05:05 +020073 resume_swap_area, as defined in kernel/power/suspend_ioctls.h,
74 containing the resume device specification and the offset); for swap
75 partitions the offset is always 0, but it is different from zero for
76 swap files (see Documentation/swsusp-and-swap-files.txt for details).
Rafael J. Wysockibf73bae2006-12-06 20:34:16 -080077
Rafael J. Wysockieb57c1c2007-10-26 01:01:10 +020078SNAPSHOT_PLATFORM_SUPPORT - enable/disable the hibernation platform support,
79 depending on the argument value (enable, if the argument is nonzero)
80
81SNAPSHOT_POWER_OFF - make the kernel transition the system to the hibernation
82 state (eg. ACPI S4) using the platform (eg. ACPI) driver
83
Rafael J. Wysockibf73bae2006-12-06 20:34:16 -080084SNAPSHOT_S2RAM - suspend to RAM; using this call causes the kernel to
85 immediately enter the suspend-to-RAM state, so this call must always
86 be preceded by the SNAPSHOT_FREEZE call and it is also necessary
87 to use the SNAPSHOT_UNFREEZE call after the system wakes up. This call
88 is needed to implement the suspend-to-both mechanism in which the
89 suspend image is first created, as though the system had been suspended
90 to disk, and then the system is suspended to RAM (this makes it possible
91 to resume the system from RAM if there's enough battery power or restore
92 its state on the basis of the saved suspend image otherwise)
93
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -080094The device's read() operation can be used to transfer the snapshot image from
95the kernel. It has the following limitations:
96- you cannot read() more than one virtual memory page at a time
97- read()s accross page boundaries are impossible (ie. if ypu read() 1/2 of
98 a page in the previous call, you will only be able to read()
99 _at_ _most_ 1/2 of the page in the next call)
100
101The device's write() operation is used for uploading the system memory snapshot
102into the kernel. It has the same limitations as the read() operation.
103
104The release() operation frees all memory allocated for the snapshot image
Rafael J. Wysockicc5d2072007-10-26 01:03:33 +0200105and all swap pages allocated with SNAPSHOT_ALLOC_SWAP_PAGE (if any).
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -0800106Thus it is not necessary to use either SNAPSHOT_FREE or
107SNAPSHOT_FREE_SWAP_PAGES before closing the device (in fact it will also
108unfreeze user space processes frozen by SNAPSHOT_UNFREEZE if they are
109still frozen when the device is being closed).
110
111Currently it is assumed that the userland utilities reading/writing the
Rafael J. Wysockibf73bae2006-12-06 20:34:16 -0800112snapshot image from/to the kernel will use a swap parition, called the resume
113partition, or a swap file as storage space (if a swap file is used, the resume
114partition is the partition that holds this file). However, this is not really
115required, as they can use, for example, a special (blank) suspend partition or
Rafael J. Wysockicc5d2072007-10-26 01:03:33 +0200116a file on a partition that is unmounted before SNAPSHOT_CREATE_IMAGE and
Rafael J. Wysockibf73bae2006-12-06 20:34:16 -0800117mounted afterwards.
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -0800118
Rafael J. Wysockiaf508b32007-10-26 00:59:31 +0200119These utilities MUST NOT make any assumptions regarding the ordering of
120data within the snapshot image. The contents of the image are entirely owned
121by the kernel and its structure may be changed in future kernel releases.
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -0800122
123The snapshot image MUST be written to the kernel unaltered (ie. all of the image
124data, metadata and header MUST be written in _exactly_ the same amount, form
125and order in which they have been read). Otherwise, the behavior of the
126resumed system may be totally unpredictable.
127
128While executing SNAPSHOT_ATOMIC_RESTORE the kernel checks if the
129structure of the snapshot image is consistent with the information stored
130in the image header. If any inconsistencies are detected,
131SNAPSHOT_ATOMIC_RESTORE will not succeed. Still, this is not a fool-proof
132mechanism and the userland utilities using the interface SHOULD use additional
133means, such as checksums, to ensure the integrity of the snapshot image.
134
135The suspending and resuming utilities MUST lock themselves in memory,
136preferrably using mlockall(), before calling SNAPSHOT_FREEZE.
137
Rafael J. Wysockicc5d2072007-10-26 01:03:33 +0200138The suspending utility MUST check the value stored by SNAPSHOT_CREATE_IMAGE
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -0800139in the memory location pointed to by the last argument of ioctl() and proceed
140in accordance with it:
1411. If the value is 1 (ie. the system memory snapshot has just been
142 created and the system is ready for saving it):
143 (a) The suspending utility MUST NOT close the snapshot device
144 _unless_ the whole suspend procedure is to be cancelled, in
145 which case, if the snapshot image has already been saved, the
146 suspending utility SHOULD destroy it, preferrably by zapping
147 its header. If the suspend is not to be cancelled, the
148 system MUST be powered off or rebooted after the snapshot
149 image has been saved.
150 (b) The suspending utility SHOULD NOT attempt to perform any
151 file system operations (including reads) on the file systems
Rafael J. Wysockicc5d2072007-10-26 01:03:33 +0200152 that were mounted before SNAPSHOT_CREATE_IMAGE has been
Rafael J. Wysocki6e1819d2006-03-23 03:00:03 -0800153 called. However, it MAY mount a file system that was not
154 mounted at that time and perform some operations on it (eg.
155 use it for saving the image).
1562. If the value is 0 (ie. the system state has just been restored from
157 the snapshot image), the suspending utility MUST close the snapshot
158 device. Afterwards it will be treated as a regular userland process,
159 so it need not exit.
160
161The resuming utility SHOULD NOT attempt to mount any file systems that could
162be mounted before suspend and SHOULD NOT attempt to perform any operations
163involving such file systems.
164
165For details, please refer to the source code.